Closed tobias-tengler closed 3 months ago
So good! Thank you!
@captbaritone has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.
Still working on landing this. It looks like it's triggering a number of errors internally. Trying to figure out if this is because we're catching real errors (and if so, how we can resolve them), or if there's something else going on here.
Is this still blocked internally?
Sorry, haven't been able to revisit. Adding to my todo list for next week.
So we have many infractions in our codebaese today, and I don't see an obvious way to "fix" these issues at scale. Maybe we need a mechanism to selectively ignore this warning for existing infractions? Perhaps we could leverage our feature flag infra here?
Also, would be cool to suggest types with similar names in this error, I think we have examples of that elsewhere in the codebase.
@captbaritone Added object type suggestions and a feature flag to disable the validation :)
@captbaritone has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.
@captbaritone I fixed the conflicts :)
@captbaritone has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.
@captbaritone merged this pull request in facebook/relay@53b24361bb4d761411474508bd08f61fedb6b1cc.
This pull request has been reverted by 43bd9bedb2b05920e61c093c3311f68bfd910fd8.
@captbaritone why was this reverted? Internal reasons?
@tobias-tengler Yeah, internal reasons. Just some internal miscommunication with getting the feature flag set. I have an internal diff open to re-land.
Currently you can input any string for the
edgeTypeName
argument on the@prependNode
and@appendNode
directives.This change enforces that the value you supply points to an existing object type within the schema.